Revised assignment of the electron spin resonance spectrum obtained

ACS Legacy Archive. Cite this:J. Org. Chem. 44, 20, 3515-3519 ... Tristram Chivers. Chemical Reviews 1985 85 (5), 341-365. Abstract | PDF | PDF w/ Lin...
5 downloads 0 Views 599KB Size
J. Org. Chem., Vol. 44, No. 20, 1979

Photolysis of Tetrasulfur Tetranitride-Bis(norbornene) evaporated in vacuo. The residue (0.7 g) was purified by TLC on silica gel using hexcane-ethyl acetate (2:3) as the developing solvent. After crystallization from dichloromethane-ether compound 2g had: mp 231-233 "C dec; ["ID +58O (c 0.30, CHCI,); UV (MeOH) 220 nm ( E 316); IR (KBr) 3560,3445, (br), 1778 cm-'; mass spectrum m / e 390 (M'). Anal. Calcd for C231-IA40s.0.25Hz0: C, 69.93; H, 8.75. Found: .~ .. . C, 69.97; H, 8.61. 18,20-0xido-20,22-dihydrodigitoxigenin Acetate (2i) by Acetylation of 2g. A solution of 2g (0.20 g) in pyridine (5 mL) and acetic anhydride (2 mL) was left at room temperature for 20 h. Water was added to the solution, and the solid was collected by filtration and purified by TLC on silica gel [hexane-ethyl acetate (3:7)]. Crystallization of the material thus obtained from ethyl acetate gave an imalytical specimen: mp 234-236 "C dec (softens a t 223 "C); [.ID +19" (c 0.31, CHCl3); UV (MeOH) 219 nm ( E 170); IR (CHCL,) 3535, 1792, 1732, 1625 (br, w); mass spectrum m / e 432 (M'). Anal. Calcd for Cz6H,0s: C, 69.42; H, 8.39. Found: C, 69.48; H, 8.47. 18-Acetoxydigitoxigenin 3-Acetate (3b). A solution of the acetate 2i (0.490 g) in triethylamine (50 mL) containing acetic anhydride (10 mL) and 4-dimethylaminopyridine (0.1 g) was heated at reflux temperature for 18 h. The solution was poured into ice water and extracted with ethyl acetate. The extract was washed with water to neutrality, dried, and evaporated in vacuo. The residue was purified by TLC on silica gel [ethyl acetatehexane (3:2)], and thle material thus obtained (0.27 g) was subjected to another purification by TLC. The amorphous powder (0.145 g, 27%) was pure: [a],, + 48" (c 0.224, CHClJ; UV (MeOH) 219 nm (c 14 100); IR (KBr) 3495,1785 (m), 1743,1629 (w) cm-'; NMR (CDCl,) 6 0.96 (s, 3 H, 19-CHJ, 2.01 (s, 3 H, CH3C02), 2.04 (s. 3 H, CHSCOZ), 3.96 (d, 1 H, JAB= 11.7 Hz, H - 1 8 ~ )4.44 , (d,

3515

1 H, J A B = 1.7 Hz, H-l8B), 4.89 (t, 2 H, 521,22 = 2 Hz, 21-CHJ, 5.08 (m, 1 H, H-3), 6.01 (t, 1 H, 521,22 = 2 Hz, 21-CH2);mass spectrum m / e 474 (M'). Anal. Calcd for C27H3807:C, 68.33; H, 8.07. Found: C, 68.22; H, 8.15. Compound 3b was also obtained in the absence of 4-dimethylaminopyridine [41% (100% purity)],but the reaction time was 72 h. 18,20-0xido-20,22-dihydrodigitoxigenin3-Acetate from 3b. A solution of 3b (1.01 g) in anhydrous methanol (80 mL) containing DBN (0.8 mL) was left at room temperature for 18 h. Saturated sodium chloride solution was added, and the product was extracted into ethyl acetate. The extract was washed to neutrality with water, dried over sodium sulfate, and evaporated in vacuo. The residue [0.84 g, 89%; ["ID +21° (c 0.255, CHCl,); UV (MeOH) 218 nm ( E 490)] was further purified by TLC on silica gel [ethyl acetate-hexane (2:3)] and by crystallization from ethyl a c e t a h t h e r . T h e material obtained in this way had mp 214-217 "c dec (softens at 205 "c);["ID +50° (c 0.26, CHCl3); w (MeOH) 215 nm ( E 280); IR (CHCI,) 3545, 1790, 1732 cm-'. Anal. Calcd for C25H3s06:C, 69.42; H, 8.39. Found: C, 69.59; H, 8.43. A small amount (0.075 g) of the alcohol 2g was also obtained in this reaction. Registry No. la, 466-07-9; lb, 25633-33-4; Id, 25633-35-6; le, 6859-20-7;le acetate, 71129-70-9 (R)-2a,71129-71-0; (S)-2a,71183-76-1; (R)-2b,71183-77-2; (S)-2b, 71129-72-1; ( R ) - ~ c71129-73-2; , (S)-~C, 71183-78-3;(R)-2d,71129-74-3;(S)-2d,71183-79-4;(R)-2e,71129-75-4; (S)-2e, 71183-80-7; (R)-2f,71129-76-5; (S)-2f, 71183-81-8; (R)-2g, 71129-77-6;(S)-2g,71183-82-9;(R)-2h,71129-78-7;(S)-2h,71183-83-0; (R)-2i,71129-79-8; (S)-2i,71183-84-1;3b, 71129-80-1;3c, 808-19-5.

Revised Assignment of the Electron Spin Resonance Spectrum Obtained by Photolysis of Tetrasulfur Tetranitride-Bis(n0rbornene)' S. Rolfe,*" D. Griller,2 K. U. Ingold,*2 a n d L. H. Sutcliffe4 Diui.sion of Chemistry, National Research Council of Canada, Ottawa, Canada K I A OR6, and Donnan Chemistry Laboratories, The University, Liverpool, England L69 3BX

Received February 6, 1979 Isotopic labeling with 33Shas been used to show that the radical obtained by photolysis of tetrasulfur tetranit ride-bis(norbornene) and by the thermal reaction of tetrasulfur dinitride with norbornene is exo-1,2norbornyl-1',3',2'-dithiazolidin-2'-yl. Several analogous radicals have been prepared and it is suggested that S4N4 adducts with unsaturated hydrocarbons may provide a starting point for the synthesis of compounds containing the I ,3,2-dithiazolidine ring system. Becke-Goehring a n d Schlafer5 first reported that tetrasulfur tetranitride6 forms 1:2 adducts with norbornene and norbornadiene. T h e structure they proposed for these adducts5 a n d alternative structures suggested by Gleiter7 and b y B r i n k m a n and AllenB have been shown t o b e incorrect."" For example, X-ray crystallographic studiesgJO of the norbornadiene a d d u c t have shown t h a t a C=C (1) Issued as K.R.C.C. No. 17740. (2) N.R.C.C. (3) N.R.C.C. co-op stu'dent, 1978. (4) IJniversity (of Liverpool. ( 5 ) Becke-Goehring, M.; Schlafer, D. Z. Anorg. Allg. Chem. 1968,356, 234. (6) For a review of sulfur nitrides see: Becke-Goehring, M. Prog. Inorg. Chern. 1959, 1, 207. (7) Gleiter, R. ,I. Chem. SOC.A 1970, 3174. (8) Brinkman, LM.R.; Allen, C. W. J.A m . Chem. SOC.1972,94, 1550. (9) Griffin, A. M.; Sheidrick, G. M. Acta Crystallogr., Sect. E 1975, 31, 895. (10) Ertl, G.; Weiss, J. Z. Anorg. Allg. Chem. 1976, 420, 155. (11) Mock, W. L.; Mehrotra, I. J . Chern. SOC.,Chem. Comrnun. 1976, 123.

0022-3263/79/ 1944-3515$01.00/0

double bond in each olefin adds across a n S-N-S unit to give a five-membered C-S-N-S-C ring. Analogous structures a p p e a r t o be formed with o t h e r strained olefins.l' T h u s norbornene will yield a n a d d u c t having structure 1.

Sutcliffe a n d Brinkman12J3 have recently reported t h a t photolysis of 1 in degassed solutions yields an extremely persistent radical X. which has a lifetime of a year or more (eq 1). Compounds analogous t o 1 also gave persistent

1

-x. hv

(12) Sutcliffe, L. H.; Brinkman, M. R. British Patent 1445968, 1976. (13) Brinkman, M. R.; Sutcliffe, L. H. J . Mugn. Reson. 1977,28, 263.

0 1979 American Chemical Society

Rolfe et al.

3516 J . Org. Chem., Vol. 44, No. 20, 1979

Table I. Selected ESR Parameters for Some Sulfur-Nitrogen Radicalsa radical

g

aN,G

ref

X. Et,NS.

2.0063 2.0156

13.0 10.7

13 19

L (CH, ),Ns. Et,NSO

2.0154 2.0060

10.7 6.1

19 19

Gii2NSO Me ,CN SC,H Me,CNSCMe, Me,CC(O)NSC,H, C,H,NSC,H,

2.0061 2.0069 2.0073 2.0082 2.0059

6.3 11.7 12.4 7.5 9.6

19 21 22 23 24, 25

2.0081

8.1

L

-Vacuum

26

Me,CNSN(CMe,)SiMe, 2.0064 12.0b 27 Me,CNS( 0 ) M e 2.0044 8.4 19, 28 Me ,CNS( 0,)Me 2.0044 12.9 29, 30 C,H,SNSC,H, 2.0082c 11.4c 31, 32 a Several of these radicals show additional hyperfine splitting by hydrogen. Splitting by the trivalent nitroUnder the same conditions as those gen is not resolved. of ref 32 (benzene at room temperature) we obtain g = 2.0076, a N = 11.4 G , and a J J S( 2 33S)= 3.71 G. The 33S hfs was obtained from satellite lines due to 33Sin natural abundance.

radicals. The ESR spectra of all these radicals were very similar.12J3 For example, in chloroform at room temperature X- itself shows hyperfine splitting (hfs) by one nitrogen and two equivalent hydrogens, the parameters beingI3 uN = 12.96 G, uH(2 H) = 3.48 G, g = 2.0063. Radical X. was a~signed’~,’~ the 1-aziridylthiyl structure 2.

H

2

Radical X. and related radicals are interesting because of their persistence,14J5though structure 2 appears rather improbable. In the first place, a massive rearrangement would be required to obtain 2 from 1. Secondly, ESR work on acyclic and cyclic (dialky1amino)thiyl radicals16-19has yielded ESR parameters for these radicals which are very different from those reported13for X. (cf. the g values listed in Table I). Our doubts about structure 2 were further heightened by the fact that three-membered-ring nitroxides have never been observed. In fact, an investigation into the reaction of the aziridinyl radical with oxygen showed that ethylene (and NO) was formed very rapidly.20 The g values and uN values for X. and a variety of other radicals containing sulfur and nitrogen16-19~21-32 are com(14) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1976, 9, 13. (15) Mendenhall, G. D.; Griller, D.; Ingold, K. U. Chem. Brit. 1974,10, 248. (16) Maillard, B.; Ingold, K. U. J . Am. Chem. SOC.1976, 98, 520. (17) Bennett, J. E.; Sieper, H.; Tavs, P. Tetrahedron 1967,23, 1697. (18) Danen, W.C.;Newkirk, D. 0. J . Am. Chem. SOC. 1976,98, 516. (19) Baban, J. A,; Roberts, B. P. J . Chem. SOC., Perkin Trans. 2 1978, 678. (20) Maeda, Y.;Ingold, K. U. J . Am. Chem. SOC.1979, 101, 837. (21) Miura, Y.;Asada, H.; Kinoshita, M. Bull. Chem. SOC.Jpn. 1977, 50,1855. (22) Miura, Y.;Asada, H.; Kinoshita, M. Chem. Lett. 1978, 1085.

U

F i g u r e 1. Apparatus used to prepare 33S-labeledX..

pared in Table I. The parameters for X. correspond most closely with those found for the thioamino class of radicals RNSR’. However, just as with 2, a valid objection to this type of structure is that considerable molecular reorganization would be required if R’ is an alkyl group. Moreover, there is no obvious structure for X. which would give hyperfine splitting by two equivalent protons. In addition, it has been reported that certain thioamino radical^^'^^^ (though not are (like certain related r a d i ~ a l s ~ ~ !not ~ ~ sensitive , ~ ’ , ~ ~ to ) atmospheric oxygen. In contrast, X. is immediately destroyed by oxygen13 (vide infra). When it became obvious that the structure of X. could not be inferred from literature data, more detailed examinations of its ESR spectrum were undertaken independently in Ottawa and in Liverpo01.~~The two studies are reported together in this paper.33 They reveal that X. contains two magnetically equivalent sulfur atoms, and we assign the exo-1,2-norbornyl-l’,3’,2’-dithiazolidin-2’-yl structure 3 to X-.33Experiments with certain other olefins and acetylenes plus S4N4have served to confirm the cyclic nature of X- and related radicals. (23) Miura, Y.;Katsura, Y.; Kinoshita, M. Bull. Chem. SOC. Jpn. 1978, 51, 3004.

(24) Miura, Y.;Kinoshita, M. Bull. Chem. SOC. J p n . 1977, 50, 1142. (25) Miura, Y.;Katsura, Y.; Kinoshita, M. Chem. Lett. 1977, 409. (26) Mayer, R.; Domschke, G.; Bleisch, S.; Bartl, A. Tetrahedron Lett. 1978, 4003. (27) Brunton, G.;Taylor, J. F.; Ingold, K. U. J . Am. Chem. SOC.1976, 98, 4879. (28) Miura, Y.;Nakamura, Y.; Kinoshita, M. Chem. Lett. 1978, 521. (29) Zomer, G.; Engberts, J. B. F. N. Tetrahedron Lett. 1977, 3901. (30) See also: Teeninga, H.; Engberts, J. B. F. N. Recl. Trao. Chim. Pays-Bas 1978,97,59. Miura, Y.; Nakamura, Y.; Kinoshita, M. Bull. Chem. SOC. Jpn. 1978, 51, 947. (31) B?on, D. H.R.; Blair, I. A,; Magnus, P. D.; Norris, R. K. J. Chem. SOC., Perkin Trans. 1 1973, 1031. (32) Miura, Y.;Makita, N.; Kinoshita, M. Bull. Chem. SOC.Jpn. 1977, 50, 482. (33) In this connection it must be added that a previous attempt to determine the number of sulfur atoms in the X. analogue derived from 3’,6’-dimethoxybenzonorbornadiene by using %S labeling led to the incorrect conclusion that the radical contained only a single sulfur atom.34 This conclusion arose from difficulties in the interpretation of the ESR spectrum which can be attributed to asymmetric line broadening with this much larger radical. (34) Fairhurst, S.A,; McIlwaine, M. R.; Sutcliffe, L. H. J . Magn. Reson. 1979, 35, 121.

J . Org. Chem., Vol. 4 4 , No. 20, 1979 3517

Photolysis of Tetrasulfur Tetranitride-Bis(norbornene)

H

3

Experimental Section Materials. Norbornene and the other olefins and acetylenes were commercial materials that were used as received. exo,exo-5,6-Dideuterionorbornenewas a gift from Professor A. P. Marchand. Tetrasulfur tetranitride was prepared by the method of Jolly and Becke-Goehring.35s:ffiThis compound together with tetrasulfur dinitride, which is a major b y p r ~ d u c twas , ~ ~also prepared with sulfur enriched with by following the general procedure of Lipp et al.,3s in the high-vacuum system shown in Figure 1. About 3 mg of =S (59.15 atom % of 33S,Oak Ridge National Laboratory) was reacted in flask A. with gaseous chlorine. After removal of the excess chlorine under vacuum,%the mixture of sulfur chlorides was distilled under vacuum into flask B. About 8.5 mL of CCl, was condensed onto the sulfur chlorides, and gaseous NH3 was then added to this flask. It was mixed with the CC14 solution by means of a magnetic stirrer. After about 10 min of mixing at room temperature (longer limes destroyed the sulfur nitrides), the reaction mixture was tipped through the glass wool filter into the ESR tube. The NH3 and CCl, were removed under vacuum and then norbornene and CHC1, were distilled into the tube, which was finally sealed off under vacuum. Samples prepared in this m y showed a strong ESR signal due to X. (or 33S-labeledX.) without photolysis. This is because there is a direct reaction ~ reaction between the olefin and the S4Nz " b y p r ~ d u c t " . ~This yields X. efficiently even at temperatures as low as -30 "C. The sulfur nitrides labeled with (48.62atom % of %) which were reacted with the dideuterionorbornene were prepared by a somewhat different procedure.37 Dibenzenesulfenamide, (C6H5S)2NH,was prepared by the method of Lecher et ,31.39 ESR Procedures. The adduct 1 was generally prepared in situ by addition of purified S4N4,norbornene, and CHC13 to an ESR tube. The samples were degassed and sealed under vacuum. At room temperature there was little or no ESR signal until the samples were photolyzed or heated to 40-60 "C. In contrast, a strong ESR signal was obtained immediately and quite without photolysis if a degassed mixture of S4N4and S4Nz in CHC1, or CCl, was added to the norbornene. Experiments with purified S4N4 and other unsaturated compounds were carried out in the same general way, except in the case of ethylene which acted as both solvent and reagent and for which the temperature was kept at or below -80 "C. Spectra were recorded on Varian E 104 (Ottawa) and E 4 (Liverpool) EPR spectrometers. The precise field positions and microwave frequencies of some of the lines were measured directly. An exact solution of the isotropic Hamiltonian was then used to compute the ESR spectroscopic parameters. Experiments to determine the susceptibility of X. toward O2 were performed by generating the radicals photochemically a t room temperature under a slow stream of oxygen-free N2which issued as a stream of bubbles from a fine capillary placed in the ESR sample tube. Since X. is persistent, the light could be cut off with no loss of radical concentration. The nitrogen capillary was withdrawn and replaced by another capillary preadjusted to provide a stream of small O2 bubbles. The X. radicals were destroyed in a few seconds and no other ESR signals appeared. Similar experiments n which (C6H5S12NHwas photolyzed in benzene or CHC13to produce the purplem dibenzenesulfenamidyl (35) Jolly, W. L.; Becke-Goehring, M. Inorg. C h e m . 1962, 1 , 76. (36) For an alternative synthesis see: Villena-Blanco, M.; Jolly, W. L. Inorg. Synth. 1976, 9, 98. (37) Apter, S.; Carruthers, M.; Sutcliffe, L. H. Inorg. Chim. Acta 1978, 31, L455.(38) Lipp, S. A,; Chang, J. J.; Jolly, W. L. Inorg. Chem. 1970, 9, 1970. (39) Lecher, H.; Holschneider, F.; Koberle, K.; Speer, W.; Stocklin, P. Ber. Dtsch. Cherrl. Ges. 1925, 58, 409.

I

I

I

i

'I1

10 G

1

I

J

Figure 2. Experimental ESR spectrum of unlabeled X. in CHC1, at room temperature. Table 11. ESR Parameters for Radicals Generated from Sulfur Nitrides and Some Unsaturated Hydrocarbons and for the Dibenzenesulfenamidyl Radicala radical solvent g aN aH a"s CHC1, 2.0064 13.11 3.49 (2) 2.89 (2)

a s \ N/,

'm:>.

CC1,

2.0064 13.09 3.41 (2) 2.81 (2)

CCI,

2.0065 12.75 3.09 (2) 2.76d 0.32 (1)'

D

OMe

N.* OMe -

CHCl, 2.0061 12.3 3.7 (2) 3.1 (1)

mi).'

CHC1, 2.0063 10.8 1.10 (1) CHC1, 2.0063 10.8

N :$':.

u : ; ; x ; ; N .

CHC1, 2.0062 11.5

3.71 (2) a Hyperfine splittings were measured a t room temperature and are given in gauss. Numbers in parentheses refer Data are from ref 34. to number of equivalent nuclei. Assigned34t o the 7-anti hydrogen. Originally assigned34t o one 33S;see footnote 33. ' Spectra were weak and the radical was not persistent; see footnote 43.

radical, (C8H&N., confirmed r e p o r t ~ that ~ ~ sthis ~ ~radical is not destroyed by 02. Like many other persistent r a d i c a l ~ , ' ~ X. J ~ ,(and ~ ~ related radicals) undergoes a reversible dimerization at low temperatures.42 (40) Lecher, H.; Koberle, K.; Stiicklin, P. Ber. Dtsch. Chern. Ges. 1925, 58, 423.

(41) Adamic, K.; Bowman, D. F.; Gillan, T.; Ingold, K. U. J . Am. C h e m . SOC.1971, 93, 903 and subsequent papers in this series.

3518

J. Org. Chem., Vol. 44, No. 20, 1979

Rolfe et al.

Figure 4. Experimental ESR spectrum of 33S-(48.62 atom % of 33S)labeled dideuterio-X. in CC14 at room temperature. Figure 3. Experimental ESR spectrum of 33S-(59.15 atom % of %) labeled X. in CHC13 at room temperature.

Results and Discussion Brief UV photolysis of S4N4and norbornene in degassed chloroform directly in the cavity of an ESR spectrometer yields the characteristic nine-line spectrum of X. (see. Figure 2). The ESR parameters obtained in this work (which differ slightly from the previously reported values12J3)are recorded in Table 11. Once X. was formed, its concentration decreased on reexposure to the UV light, but when the light was again cut off, the concentration of X. rose within a few seconds, often to a higher value than before. It is clear that X. is photolabile. It would also appear that X. is not formed by direct photolysis of 1 but rather by the thermal decomposition of some short-lived intermediate. Since S4N4itself does not yield S4N2 on the intermediate in question may be the same as the initial S4N2-norbornene product, which could be formed, perhaps, by the photolytic elimination of N2 from 1. At high gain the ESR spectrum of X- showed two incompletely resolved satellite lines, one a t each end of the spectrum, separated from the outermost lines by ca. 4.3 G. The heights of these lines suggested that they might be due to a single %S atom a t natural abundance (0.76%). However, integration suggested that these lines were more probably due to two equivalent 33Satoms. The radical derived from norbornene and sulfur nitrides labeled with 33S(59.15atom % of 33S)gave the ESR spectrum shown in Figure 3. This spectrum can be analyzed and simulated on the basis that X. contains two magnetically equivalent 33Satoms with aSgS= 2.89 G-it is, of course, a composite spectrum arising from 35.0% of X- having two 33S,48.3% of X. having only one 33S,and 16.7% of unlabeled X.. Furthermore, the line width for X- containing two 33S nuclei is about 50% greater than the line width for the unlabeled X.. In CC14 the radical derived from dideuterionorbornene labeled with 33S(48.62atom % of 33S)gave the spectrum shown in Figure 4. This spectrum is much less intense but the lines are sharper than for the undeuterated radical. The spectrum can also be analyzed and simulated on the basis that X. contains two magnetically equivalent 33S. The EPR parameters obtained are given in Table 11. (42) Sutcliffe, L. H., unpublished results.

We assign X. the cyclic exo-1,2-norbornyl-l’,3’,2’-dithiazolidin-2’-yl structure 3 and radicals similar to X. a similar cyclic structure (see Table 11). This structure is logical in that it contains two equivalent sulfurs and can be formed from 1 and related adducts by a minimum of molecular reorganization, e.g., 1% 3 + N2 + 3. The potential acyclic structure 4 can be discounted for several U

U

5

4

reasons. In the first place, the g values for X. and related radicals (Table 11) are appreciably less than the g value for the acyclic dithioamino analogue 5 (see Tables I and 11). Secondly, some radicals that are obviously closely related to X. have hydrogen hyperfine splittings which are inconsistent with an acyclic structure but are quite consistent with the proposed cyclic structure. Examples include the radicals from (i) di~yclopentadiene’~ which probably has aH(l H) = 3.4 G and aH(l H) = 3.7 G, presumed to be 6,(ii) ~ t y r e n e )aH(2 ~ H) = 3.7 G, aH(l H) = 3.1 G (see Table 11),and (iii) phenyla~etylene)~ aH(l H) = 1.10 G (see Table 11).

H

6

The acyclic dithioamino radical 5 (and related species32) and the cyclic dithioamino radicals show interesting differences both in their ESR parameters and in their reactivities. We believe that these differences may best be accounted for by assuming that the dithiazolidinyl ring adopts an envelope conformation 7 in which the nitrogen does not lie in the plane defined by the two carbon and two sulfur atoms. In this conformation there will be poor overlap between the semioccupied N 2p, orbital and the p-type lone pairs on the two sulfur atoms. As a result, (43) In this case, the spectrum was very weak even under strong UV photolysis and the radicals decayed instantly when the light was cut off. In fact these radicals could not be detected by the Liverpool group, but their spectra appear, in general, to be less intense than those obtained in Ottawa (cf. Figures 3 and 4).

J . Org. Chem., Vol. 44, No. 20, 1979 3519

Electrochemical Oxidation of Enamines

trides and a variety of strained46o l e f i n ~ , ’ ~ J ~ and certain acetylenes47(see Table 11),it seems not unlikely that 1 and related compounds could serve as the starting point for the synthesis of compounds containing this unexplored heterocyclic ring system.4s Acknowledgment. L.H.S. thanks Miss S. A. Fairhurst for technical assistance and Professor A. P. Marchand (Oklahoma) for the dideuterionorbornene.

7

conjugative electron delocalization from N to S will be disfavored. The cyclic dithioamino radicals will therefore have smaller g values and smaller hyperfine splittings by 33Sthan the acyclic dithioamino radicals (as is observed; see Table 11). In addition, because the cyclic radical is less stabilized than the i~cyclicradical by conjugative electron delocalization, it is more reactive. This provides a simple explanation as to why 3 reacts with oxygen but 5 does not. The 1,3,2-dithiazolidinering is almost u n k n ~ w n .Since ~ radicals analogous to 3 can be produced from sulfur ni(44) The only example we could discover in the literature45 was SCHZCH2SNCH;g. (45) Mueller, W.H.; Dines, M. J . Heterocycl. Chem. 1969, 6 , 627.

Registry No. 1, 61501-72-2; 3, 71042-46-1; 5 , 41601-45-0; 6, 71042-47-2; S4N4,28950-34-7; norbornene, 498-66-8;exo,exo-5,6-dideuterio-1,2-norbornyl-1’,3’,2’-dithiazolidin-2’-yl, 71042-48-3; 5,8dimethoxy-4,9-methanonaphtho[2,3-d] [ 1,3,2]dithiazol-2-y1,71042-49-4; 4-phenyl-1,3,2-dithiazolidin-2-yl, 71042-50-7; 4-phenyl-1,3,2-dithiazol-2-y1, 71042-51-8; 4,5-diphenyl-1,3,2-dithiazol-2-yl, 71042-52-9; 4,5-bis(trifluoromethyl)-1,3,2-dithiazol-2-y1, 71042-53-0. (46) We failed to obtain radicals with the following olefins: ethylene, tetramethylethylene, bromotrifluoroethylene, cyclopentene, 3-ethylcyclopentene, cyclohexene,3,3,4,4-tetrafluorocyclobutene, tetramethyldene, citraconic anhydride, maleic anhydride, N-vinylimidazole, methyl methacrylate, stilbene, and thiophene. (47) We failed to obtain radicals with 1-pentyne, 3,3-dimethyl-l-butyne, and 2-butyne. (48) The S4N4adduct with trans-cyclooctene has been reduced to the dithiol with zinc and HC1.”

Electrochemical Oxidation of Enamines in the Presence of Organic Anions Toshiro Chiba,* Mitsuhiro Okimoto, Hiroshi Nagai, and Yoshiyuki Takata Department of Applied Chemistry, Kitami Institute of Technology, Kitami, Hokkaido, Japan 090

Received April 18, 1979

Anodic oxidations of enamines were examined in the presence of organic anions derived from methyl acetoacetate, acetylacetone, and dimethyl malonate. Enamines are very readily oxidized at a platinum anode and are substituted by the anionic substrates. The produced enamine intermediates are hydrolyzed with dilute hydrochloric acid to the corresponding a-substituted ketones. By treatment with more concentrated hydrochloric acid, some of them are converted to furan derivatives involving an intramolecular condensation. Enamines are very useful intermediates for the formation of ketone derivatives, and the chemical reaction of enamines has been extensively studied by various investigat0rs.l Although the anodic behavior of diaminoalkenes has been examined by cyclic polarographic and ESR spectroscopic methods,2 only a few reports seem to be available on the ielectrochemical oxidation of enamines from a synthetic point of view. Recently, Shono and his co-workers showed that the anodic oxidation of enamines in methanol containing sodium methoxide gives rise to methoxylation at the &carbon atom to the n i t r ~ g e n . On ~ the other hand, the most common organic anions derived from P-dicarbonyl compounds are well-known to readily undergo electrochemical oxidation to form radicals which can add to the double bond of 01efins.~ In our laboratory, the electrochemical reaction of enamines with these anions (1) See, for example, G.Stork, A. Brizzolara, H. Landesman, and R. Terrell, J . Am. Chem. SOC.,85, 207 (1963);A. G. Cook, Ed., “Enamines”, Marcel Dekker, New York, N.Y., 1969. (2) J. M. Fritsch, H. Weingarten, and J. D. Wilson, J . Am. Chem. SOC., 92, 4038 (1970); ,J. M. Fritsch and H. Weingarten, ibid., 90, 793 (1968); K. Kuwata and I). H. Geske, ibid., 86, 2101 (1964). (3) T. Shono, Y. Matsumura, H. Hamaguchi, T. Imanishi, and K. Yoshida, Bull. Chem. Soc. Jpn., 51, 2179 (1978). (4) H. Schafer, Chem.-lag.-Tech., 42,164 (1970);H. Schafer, Angeu. Chem., 82, 134 (1970); H,Schafer and A. Aiazrak, ibid., 80, 485 (1968).

Scheme I -‘N’

.-..

HA NoOMs-MeOH

Re

R1

1

a b c d e f g

&A Rl

R2

2, A = CH(COCH,)COOCH, 3, A = CH(COCH,), 4, A = CH(COOCH,),

morpholino morpholino morpholino piperidino piperidino pyrrolidino

(CH,), CH,, CH, (CH,), (CH,), (CH,), (CH,),

pyrrolidino (CH2)

was attempted at a platinum anode, and the synthetic application of this reaction was examined,

Results Prior to the preparative studies, current-potential measurements were carried out in methanol containing sodium in the presence and absence Of enamines at a platinum anode.

0022-3263/79/ 1944-3519$01.00/0 0 1979 American Chemical Society